24 J
.P. Manderson et al. J. Exp. Mar. Biol. Ecol. 251 2000 17 –39
Fig. 2. Spatial and temporal patterns of summer flounder catch per unit effort in trammel nets 0 CPUE5no
21
fish h 6S.E., and size distribution in the Navesink River. Station numbers are indicated on the abscissa.
deep Table 1. Predators were isolated and prey acclimated within the tanks as described above. Ten prey were exposed to each predator for 4 h and the trials were
videotaped continuously. Each 4 h videotape was analyzed to quantify: 1 the number of attacks; 2 method of attack lie-in-wait or active stalking; 3 whether prey were
visible and unburied and or; 4 moving prior to the attack; and 5 location of attacks bottom or water column.
3. Results
3.1. Field studies 3.1.1. Summer flounder size and distribution
Summer flounder collected in trammel nets ranged is size from 252 to 648 mm TL Fig. 2. The size distributions of fish were similar in June and July n 580, Median
TL5359 mm; two sample Kolmogorov–Smirnov test, KS50.23, P 50.21. Although too few individuals were collected in August to permit analysis, the size distribution also
appeared to be similar n 515, Median TL5336, 252–556 mm. Summer flounder were consistently abundant at the easternmost station Sta. 1; Fig. 2. Although fish were also
J .P. Manderson et al. J. Exp. Mar. Biol. Ecol. 251 2000 17 –39
25 Table 2
Contribution to total prey volume by vol. and occurrence for prey of summer flounder collected in the Navesink river containing stomach contents n 555
] Prey species
x 0 by vol. Occurrence
S.E. n
Sand shrimp 30.465.7
45 25 Crangon septemspinosa
Winter flounder 21.565.1
27 15 Pseudopleuronectes americanus
Blue crab 19.365.2
22 12 Callinectes sapidus
Mysids 5.463.0
11 6 Atlantic silversides
5.062.7 7 4
Menidia menidia Grass shrimp
2.761.9 11 6
Palaemonetes spp Unidentified fish
4.362.3 5 4
Atlantic menhaden 4.762.7
5 3 Brevortia tyrannus
Lady crab 3.662.5
4 2 Ovalipes ocellatus
Northern pipefish 0.861.2
4 2 Syngnathus fuscus
Other 0.360.6
1
collected at stations upstream Sta. 3–5, catches were lower and relatively few fish were collected in the upper river in August.
3.1.2. Dietary patterns Fifty-eight percent of the summer flounder collected n 595 contained prey and
stomach fullness was not correlated with environmental variables Spearmans r, P.
0.05. Sand shrimp and winter flounder were the dominant prey Table 2. Individual ]
] predators consumed as many as 27 shrimp x 55.3 and 11 winter flounder x 52.7.
Blue crabs Callinectes sapidus, mysids, and grass shrimp Palaemonetes spp. were also relatively common. Most of the prey were consumed whole.
The diets of the predators changed through time Fig. 3. Sand shrimp and winter flounder were important in June and July, but absent from diets in August. Other fishes
Menidia menidia, Brevoortia tyrannus, Syngnathus fuscus and Gobiosoma spp. and blue crabs were dominant prey in August.
Spatial variation in dietary composition only occurred with respect to sand shrimp prey River km versus shrimp percent of stomach volume, SV: Spearmans
r 5 20.36, P ,0.01. Frequency of occurrence F and percent stomach volume for shrimp were
] higher for predators collected in the upper river Sta. 4–6; x SV64, F 77, n 516
] than in the lower river Sta. 1 and 2; x SV37, F 54, n 521.
3.1.3. Prey size and predator gape size Predators consumed sand shrimp ranging from 9 to 48 mm TL n 5109; 2.4–14.4 of
predator TL and winter flounder from 24 to 67 mm TL n 560, median TL532, 6–19
26 J
.P. Manderson et al. J. Exp. Mar. Biol. Ecol. 251 2000 17 –39
Fig. 3. Temporal patterns in mean percent stomach volume 6S.E. for important prey consumed by summer flounder in the Navesink River.
of predator TL. We found no evidence for significant predator–prey body size relationships as the slopes of quantile regressions estimating maximum, minimum, and
median prey size were not different from 0 P .0.34; quantiles: winter flounder; 20th, 80th and 50th; sand shrimp, 90th, 10th and 50th. Winter flounder consumed in the field
were smaller than maxima defined by summer flounder gape dimensions Fig. 4 and Table 3.
3.2. Laboratory experiments 3.2.1. Prey selection
Summer flounder consistently selected demersal winter flounder over pelagic fish Atlantic silversides and benthic invertebrate sand shrimp prey at all ratios Fig. 5 and
Table 4. Chesson’s a values for winter flounder were significantly greater than 0.5
P 0.02 in five of six prey type prey ratio combinations Table 4. Selection for winter flounder was not statistically significant when Atlantic silversides were offered in
equal numbers i.e., 10:10. However, Chesson’s a for winter flounder averaged 0.82
S.E.50.13 in the treatment, and predators consumed more silversides in only one replicate.
3.2.2. Winter flounder body size and prey vulnerability
2
Winter flounder vulnerability to the predators increased significantly x 59.20,
P 50.002 with increasing prey body size from 37 S.E.50.06 for 20–30 mm fish
J .P. Manderson et al. J. Exp. Mar. Biol. Ecol. 251 2000 17 –39
27
Fig. 4. Scatterplot of body size relationships between summer flounder collected in the field and winter flounder found in their stomachs. Lines indicate estimated winter flounder TL mm if predator mouth height
MH9, mouth width MW9 and esophageal width EW9 determine maximum prey size. Estimates were developed from regressions for summer flounder gape Table 3 and the winter flounder total length body
depth relationship see Section 2.
10 of predator length to 88 S.E.50.03 for the 80–90 mm size class 29 of predator length; Fig. 6.
3.2.3. Effects of sediment grain size and macrophytes on prey vulnerability Winter flounder mortality resulting from summer flounder predation was not in-
fluenced by sediment grain size. In the preliminary size-dependent burial experiment, all three sizes of winter flounder buried completely in fine sand mean maximum burial
] ]
score x max. B .2.25 but could not bury in gravel x max. B 50. In coarse sand, the ]
largest fish were capable of complete burial x max. B 52, S.E.50.58, the 40–49 mm ]
fish capable of partial burial 25–50 of body covered; x max. B 51, S.E.50, and the ]
smallest fish were incapable of burying x max. B50. However, neither sediment grain
Table 3 Regressions to determine the relationship between summer flounder total length mm n 561, 252–473 mm
TL and gape dimensions mm
2
Dependent variable Effect
Coefficient S.E. T
P R
Mouth height Intercept
1.359 4.190 0.324
0.747 0.517
MH Total length
0.121 0.011 10.35
,0.001 Mouth width
Intercept 1.069 6.373
0.168 0.867
0.315 MW
Total length 0.110 0.018
6.238 ,0.001
Esophageal width Intercept
6.477 2.644 2.450
0.018 0.432
EW Total length
0.055 0.042 7.507
,0.001
28 J
.P. Manderson et al. J. Exp. Mar. Biol. Ecol. 251 2000 17 –39
Fig. 5. Proportion of winter flounder consumed by summer flounder in choice experiments in relation to the proportion of total prey offered see Table 4.
J .P. Manderson et al. J. Exp. Mar. Biol. Ecol. 251 2000 17 –39
29 Table 4
a
Results of summer flounder prey selection experiments Prey combination
Degrees of Chesson’s
a T
P Prey ratio
freedom Winter flounder
] x6S.E.
Winter flounder :Atlantic silversides
5:15 5
0.94760.039 11.465
0.000 10:10
5 0.81860.129
2.463 0.057
15:5 5
0.94460.036 12.467
0.000 Winter flounder
:sand shrimp 5:15
4 0.77760.071
3.742 0.020
10:10 3
0.93260.054 7.928
0.004 15:5
3 0.83860.057
5.952 0.009
a
Only Chesson’s a values for winter flounder are reported. a 50.5 indicate no selection.
size nor prey body size influenced winter flounder vulnerability to summer flounder predators Fig. 7 and Table 5a.
The presence of macrophytes decreased the vulnerability of winter flounder to summer flounder predators Fig. 8, Table 5b. Although prey mortality differed between
trials F 54.71, P 50.048; Table 5b, the trial3treatment interaction was insignificant
Fig. 6. Mortality probabilities 62 S.E., closed circles for winter flounder of different body sizes from summer flounder predation. Open circles indicate proportion of total prey consumed in the size classes offered.
Individual prey were offered at random to individual predators and 10 replicates for each prey size class were performed.
30 J
.P. Manderson et al. J. Exp. Mar. Biol. Ecol. 251 2000 17 –39
Fig. 7. Proportion of prey surviving 061 S.E. in experiments testing for the effects of a sediment grain size and prey body size and b macrophytes on winter flounder vulnerability to summer flounder predation.
Treatments were not significantly different P .0.05 in the sediment experiment, but were significantly different P 0.01, Fishers LSD test in the macrophyte experiment.
J .P. Manderson et al. J. Exp. Mar. Biol. Ecol. 251 2000 17 –39
31
Fig. 8. Percent of time active 061 S.E. for winter flounder and sand shrimp in 24-h experiments. Bars below abscissa indicate simulated light levels open, daylight; grey, sunrise or sunset; closed, night.
32 J
.P. Manderson et al. J. Exp. Mar. Biol. Ecol. 251 2000 17 –39 Table 5
Analysis of variance tests for the effects of a prey size total length mm and sediment grain size and b
a
vegetation type on the vulnerability of juvenile winter flounder to summer flounder predation Source of variation
Degrees of Mean
F ratio P
freedom square
a Sediment grain size
2 0.152
0.832 0.391
Prey size 2
0.160 1.003
0.375 Interaction
4 0.094
0.590 0.672
Error 45
0.154 b
Trial 1
0.147 4.710
0.048 Vegetation type
2 0.626
29.903 ,0.001
Error 14
0.031
a
Proportions of prey surviving were arcsine transformed prior to analysis. In macrophyte experiment, the trial x vegetation type interaction was not significant F 50.451, df52, P 50.65 and dropped from the final
analysis.
F 50.451, df52, P 50.65 and excluded from the final analysis. Prey survival was ]
significantly higher in eelgrass x 585, S.E.50.04 than in sea lettuce P 50.010, ]
] x 562, S.E.50.08 or sand P ,0.001, x 530, S.E.50.07; Fishers LSD test.
Survival was also higher in sea lettuce than on sand P 50.005. 3.2.4. Predator and prey behavior
] ]
Winter flounder were substantially more active x time active, x TA550, S.E.52.7 ]
than sand shrimp x TA58.9, S.E.52.7; Fig. 8. Flounder activity was highest at sunset and sunrise and the prey spent more time swimming in the water column during
] ]
night hours day x 522, S.E.53.3; night x 540, S.E.52.7. Although shrimp showed maximum activity following sunset and sunrise, most individuals remained
] ]
buried during the day and night day x TA58.5, S.E.52.6; night x TA59.6, S.E.54.2 and rarely swam in the water column. In the presence of flounder, shrimp
] remained buried x TA50.5, S.E.50.3. Although flounder activity was slightly
] depressed in the presence of shrimp x TA533, S.E.51.9, diel patterns were similar
to those observed when shrimp were absent. Summer flounder attacked exposed and active winter flounder. Of the 33 attacks
observed on videotape, most involved prey visible on the sediment surface prior to ]
attacks x 579, S.E.514, and many of the prey had been actively moving along the ]
substratum x 533, S.E.517. The predators never used a lie-in-wait attack strategy on bare sand, but stalked winter flounder, which were primarily attacked while on the
bottom 80, S.E.52.4.
4. Discussion
